-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
I used round trip poorly, it usual means A to B to A (the half second video conference delay) whereas in the 240mS statement I meant earth to sat and back since the comparison was Earth to Starlink to Earth to Earth vs Earth to Starlink to Starlink to Earth, thus the differential being the groundlink vs laser.
With that info alone, I believe you can figure out the IPv4 addresses that would be end user terminals. You should be able to do pings and then traceroutes to them. I'm thinking you might be able to tell you've hit an end user via a Starlink satellite by carefully looking at increasing ping times on the traceroute. It wouldn't be definitive proof that you are bouncing a ping off Starlink, but it would be an educated guess. Maybe others that know more about ISPs and the like would do a better job at sleuthing...
I mean, on an unloaded circuit (which Starlink would be during this alpha test period), you might even be able to see ping times go down then back up as the satellite transits above, increasing physical trip length.
Or, SpaceX will put in a filtering firewall and ruin our fun
And that is going to be a huge challenge; trying to convince rural customers who don’t realize how lousy their service is that they should switch and pay as much or more than what they are currently paying.
Krugerrand
Meow
I disagree. I know a lot of rural customers. They all know how slow their Internet is. They do have ‘city’ contacts and such. Unless of course you meant specifically Mountain Hillbillies that have never left the mountain?
It is definitely 'out there'.
Yes there are people that will pay $200/month.
There are not a lot of people that will pay $200/month.
dhanson865
Well-Known Member
I disagree, there are so many people posting on reddit in various forums that they pay over $100 a month for 2 mbit speeds, having guaranteed 100/40 with bursts above that minimum would be enough for people like that to pay $300 a month.
Personally I get 100/40 now on cable and would be willing to pay more to SpaceX than I pay to my cable company. I know that isn't going to happen soon because I'm in a metro area so population density is against me, but there are a lot of people out there that would jump on Starlink at above competitive pricing just to tell Comcast/Hughes/ATT/Dish/whoever to go jump in a lake.
Here's hoping SpaceX builds their business model around more than anecdotal internet references to halo pricing.
In any event, moving the conversation forward, for those who weren't following last week there was a crazy flurry of FCC filings/modifications for NGSO and mostly LEO constellations trying to get in before the deadline. It seems to be as much everyone trying to get their piece of spectrum and orbital real estate before SpaceX monopolizes everything (SpaceX was part of the flurry too); we'll see what actually makes it through. Notable players are Viasat and O3B with pretty sensible plans, Telesat looking to ~4x their current plans of 300 sats, SpaceX's 30k gen2, and...wait for it...from our bankrupt friends at OneWeb, 48,000 satellites.
I mean, if you were trying to improve the value of your bankrupt company that's what you'd do to, but...48k is pretty ambitious and really just an FU to SpaceX. Layering in the orbital debris conversation's likely ramifications, 1) as previously noted, pretty much anything in LEO is going to have to come down instead of the higher LEOs being able to go up and, 2) there's going to be a practical limit to number of satellites in any one altitude shell, so bigger constellations will basically have multiple shells around ideal altitudes. While the number per shell is a little fuzzy right now--obviously there's a geometrical element to the function based on altitude, but the big variable is how close you're willing to pass satellites from other planes, notably around the equator passes where relative velocities are generally the highest--its pretty much a foregone conclusion that at least some of the operators will operate their shells at near capacity.
What all that means is that all the operators are going to have to figure out how to play nice on a recurring basis, specifically when it comes to deorbiting and in many cases orbit raising, since direct inject doesn't really work. For instance, pretty much everything higher is going to ultimately have to go through the SpaceX cloud at pretty low altitudes and, like it or not, SpaceX is going to have to actively work with each and every one of those satellites, many of them twice.
e-FTW
New electron smell
If you have not read the Reddit AMA that SpaceX software Engineering did over the weekend, here is one teaser...
There is plenty more that even non-devs will enjoy.
Source: https://www.reddit.com/r/spacex/comments/gxb7j1/we_are_the_spacex_software_team_ask_us_anything/ft63jxc/
There is plenty more that even non-devs will enjoy.
So, looks like potentially good news on the FCC broadband auction front. The FCC will dole out billions of dollars in grant money over a ten year period based on an auction concluding this October. SpaceX as well as lots of terrestrial based Internet Service Providers will bid to provide Internet service to underserved, rural customers. Lowest cost/best service through a complicated formula will win grant money.
SpaceX was challenging an FCC decision which would have placed it in the same bucket as very high latency GEO satellite services.
This WSJ article didn't have the particulars, but here are some quotes:
"The Federal Communications Commission on Tuesday delayed a final decision on whether Elon Musk’s SpaceX will be able to qualify as a preferred bidder when the agency prepares to distribute up to $16 billion in funding to expand broadband service in rural areas."
"SpaceX says its technology should be considered “low-latency,” though competitors disagree. The FCC decided Tuesday to allow low-Earth-orbit satellite companies to apply for the label, FCC Chairman Ajit Pai told reporters after meeting in which the agency adopted final rules for distributing the funds."
"That wasn’t part of the agency’s initial plan, which would have barred low-Earth-orbiting technologies from qualifying for preferred “low-latency” status. The change was made after a May 29 phone call between SpaceX and FCC staff, during which the firm argued its technology “easily clears” the agency’s definition for low-latency, according to a disclosure the company filed with the FCC.
Mr. Pai said SpaceX’s application will be evaluated by FCC staff along with applications from other potential bidders in the auction. “The bottom line is of course we want to make sure that the participants in the auction, if successful, would in fact be able to deliver,” he said."
Article: FCC Delays Decision on Elon Musk’s SpaceX and Broadband Subsidies
SpaceX was challenging an FCC decision which would have placed it in the same bucket as very high latency GEO satellite services.
This WSJ article didn't have the particulars, but here are some quotes:
"The Federal Communications Commission on Tuesday delayed a final decision on whether Elon Musk’s SpaceX will be able to qualify as a preferred bidder when the agency prepares to distribute up to $16 billion in funding to expand broadband service in rural areas."
"SpaceX says its technology should be considered “low-latency,” though competitors disagree. The FCC decided Tuesday to allow low-Earth-orbit satellite companies to apply for the label, FCC Chairman Ajit Pai told reporters after meeting in which the agency adopted final rules for distributing the funds."
"That wasn’t part of the agency’s initial plan, which would have barred low-Earth-orbiting technologies from qualifying for preferred “low-latency” status. The change was made after a May 29 phone call between SpaceX and FCC staff, during which the firm argued its technology “easily clears” the agency’s definition for low-latency, according to a disclosure the company filed with the FCC.
Mr. Pai said SpaceX’s application will be evaluated by FCC staff along with applications from other potential bidders in the auction. “The bottom line is of course we want to make sure that the participants in the auction, if successful, would in fact be able to deliver,” he said."
Article: FCC Delays Decision on Elon Musk’s SpaceX and Broadband Subsidies
a) no freaking way. Most ruralites don't have bags of money to throw around. Sliding laterally in monthly cost of $100 while upping speed, in exchange for sucky latency, *might* play. $300, definitely not.
b) those who have cable access (vs merely service) should be prohibited by FCC from accessing sat Internet services as part of the spectrum license. Even if you can flash your Lifetime Membership to the Elon Musk Fanboy Club, you have access and loading up a satellite network unnecessarily would be a total dick move.
I think everyone is missing something really big.
Current rural offerings, with the possible exception of local Wireless ISPs (where they exist), are not worth "sharing".
With StarLink, if you can "see" 2, 3, 5 neighbors, it is very much worth sharing.
With the cost and easyeof setup for UBNT and Mikrotik point to point, and point to multipoint line of sight wireless equipment, I can EASILY see small groups of neighbors getting together, getting StarLink service (even at $300-$500/month), and sharing it locally.
Here are a couple of examples: NanoStation 5AC Loco and airMAX LiteBeam AC 5 GHz Bridge
Again, as long as you have line of sight, you can get these types of things to do 10+ miles easily. If you have a high barn on your property, and a high barn on the neighbor's property, you can probably make it work.
UBNT even has a nice site to model your link to see if it "should" work: Link
Reasonable estimate is $150 per end per connection (I am going to assume point to point), or $300 per connection total. Let's say StarLink CPE is $1000, and you have 5 neighbors that want in on it. Your looking at $2500 to start ($500 each), and then what ever monthly cost starlink ends up being divided by 5, so at $300, that would be $60/month.
-Harry
Also rural school systems. Get the better ground stations and it opens a wold of possibilities.
Mo City
Active Member
Interesting. Has SpaceX previously mentioned this concept applying to Starlink?
Maybe some people? But aggregating a number of customers on one feed has been talked about a lot here.
Its definitely plausible, but layering reality on to the concept relative to the big picture of starlink customers, it’s really an inconsequential factor. It’s easy to put on the “I’m a tech person and it’s easy” blinders, but let’s be real—the majority of people don’t want to have any part of their internet infrastructure. So...a more traditional WISP is a little more likely on a larger scale, with the requisite margins.
Of course, the concept also leads to fewer starlink direct customers and thus lower revenue, so that doesn’t help spacex...
Exactly. 99.9% of the population has no idea WTF we are talking about here. Most people conflate WiFi with their Internet service and get frustrated when their WiFi sucks, yet are paying for a 300 Mbps link or whatever.
If Starlink is too expensive for a rural customer, that is exactly what will happen. A WISP will buy a commercial service, pay more it, and then piece it out to local customers via point to point long range WiFi links using Ubiquiti hardware.
The idea of end users sharing their link via WiFi might happen here or there, but not in ANY real volume. Also, pretty much every single end user contract has a clause PROHIBITING the sharing of their service. If you want to share your service, fine, but pay the commercial rate.
dhanson865
Well-Known Member
what fantasy universe do you live in where Starlink is providing "sucky latency"? One of the biggest selling points of Starlink for rural users is it will have speeds and latency that rivals fiber for users that don't currently have fiber.
I mean, I'm the first to call out that The Free Market Isn't and I'm the first to point out the various socialist preferences ALL Americans have (even those who pretend to be against socialism), but the above isn't a realistic scenario for the way the country operates...and it likely steps on some anti-trust or other laws to boot.
Its maybe a bit of a moo point though, as those who have "cable service" (assuming that means a decently fast and robust/stable connection) are going to find that terrestrial service is better value than Starlink anyway.
We are still waiting to read the exact FCC language regarding the Rural Broadband Auction, but Jon Brodkin of Ars advances the story here:
Ajit Pai caves to SpaceX but is still skeptical of Musk’s latency claims
Given that SpaceX isn't objecting to not being allowed to bid in the gigabit tier, I am guessing their offering, at least initially, won't be close to a gigabit. This is consistent with another article I linked to upthread where SpaceX said they would offer 100 Mbps.
Commissioner O'Rielly is correct in that the auction specs should have been technology neutral. In general, telecoms moves too fast for bidding packages to be assuming a static world. I don't know whether FCC chairman Pai either doesn't get this, is too obtuse, or is playing footsie with established telecom carriers. The reality is that winning bidders using established technology can also fall flat on their faces. You can promise to build out a fiber network (and remember you are low bidding here), and find that you can't complete the network for one reason or another, so the FCC must (and hopefully does) have mechanisms to track this. Tracking whether or not a satellite service can deliver what they promise is no different.
The NY Times article about this (not bothering to link to it) really only talked about the divisions between the two Democrat commissioners and the three Republican commissioners. In particular, they focused on the Democratic minority wanting to delay the auction so that more accurate broadband maps could be made. The auction will grant money to census blocks where no provider offers home-Internet speeds of at least 25Mbps downstream and 3Mbps upstream. How big is a census block? It's pretty small. Average census block population is 1,500 people, so trying to slice and dice a country wide auction into smaller blocks than that is counter productive, IMHO.
Anyways, Starlink looks like they got what they wanted. It'll be very interesting indeed to see what the broadband auction results are at the end of October. I don't know if SpaceX built their business case off these grants, but it certainly can't hurt. In past auctions, while the lions share of the money went to established telcos (just because they can bid for more places), a decent amount went to small projects where a small ISP would bid to usually completely pay for the construction of a small region or community gigabit fiber network.
Ajit Pai caves to SpaceX but is still skeptical of Musk’s latency claims
Given that SpaceX isn't objecting to not being allowed to bid in the gigabit tier, I am guessing their offering, at least initially, won't be close to a gigabit. This is consistent with another article I linked to upthread where SpaceX said they would offer 100 Mbps.
Commissioner O'Rielly is correct in that the auction specs should have been technology neutral. In general, telecoms moves too fast for bidding packages to be assuming a static world. I don't know whether FCC chairman Pai either doesn't get this, is too obtuse, or is playing footsie with established telecom carriers. The reality is that winning bidders using established technology can also fall flat on their faces. You can promise to build out a fiber network (and remember you are low bidding here), and find that you can't complete the network for one reason or another, so the FCC must (and hopefully does) have mechanisms to track this. Tracking whether or not a satellite service can deliver what they promise is no different.
The NY Times article about this (not bothering to link to it) really only talked about the divisions between the two Democrat commissioners and the three Republican commissioners. In particular, they focused on the Democratic minority wanting to delay the auction so that more accurate broadband maps could be made. The auction will grant money to census blocks where no provider offers home-Internet speeds of at least 25Mbps downstream and 3Mbps upstream. How big is a census block? It's pretty small. Average census block population is 1,500 people, so trying to slice and dice a country wide auction into smaller blocks than that is counter productive, IMHO.
Anyways, Starlink looks like they got what they wanted. It'll be very interesting indeed to see what the broadband auction results are at the end of October. I don't know if SpaceX built their business case off these grants, but it certainly can't hurt. In past auctions, while the lions share of the money went to established telcos (just because they can bid for more places), a decent amount went to small projects where a small ISP would bid to usually completely pay for the construction of a small region or community gigabit fiber network.
Your forgetting we are talking about Rural America here.
This is the heart of "DIY" in the US (and Canada). Maybe they won't do it themselves, but the high school student three ranches over will do it for them in a heart beat.
I am not saying having less direct customers is good for SpaceX, but having more total customers is. If the cost per terminal (CPE and service) is too high, a way around that will be found. The overall bandwidth used will not change that much, and SpaceX will still have more customers.
It may also permit WISPs to provide service where they could not get clear connectivity back to real bandwidth before.
-Harry
Similar threads
